Fix exception handling in commands calling list_repository_contents
#3968
Conversation
The commands `verdi node repo ls` and `verdi calcjob inputls/outputls` were passing the caught exception object straight into the echo call without casting it to a string first, which would incur another exception by itself. On top of that, they were catching `ValueError` exceptions, which, as far as I can tell, would never be thrown. Instead if the path does not exist a `FileNotFoundError` will be thrown. The commands now catch this instead and print a useful message. This failure path of the commands was not tested, so the bug went unnoticed. The problem of raw exception objects being passed to `echo_critical` were present in some other commands as well, which have now been fixed.
Codecov Report
@@ Coverage Diff @@
## develop #3968 +/- ##
========================================
Coverage 78.33% 78.33%
========================================
Files 461 461
Lines 34076 34076
========================================
+ Hits 26692 26694 +2
+ Misses 7384 7382 -2
Continue to review full report at Codecov.
|
| @@ -176,8 +176,8 @@ def calcjob_inputls(calcjob, path, color): | |||
|
|
|||
| try: | |||
| list_repository_contents(calcjob, path, color) | |||
| except ValueError as exception: | |||
| echo.echo_critical(exception) | |||
| except FileNotFoundError: | |||
There was a problem hiding this comment.
the only comment I have is that this might become a different exception when we move to the new repository implementation.
is there perhaps a higher-level exception that we could catch here?
There was a problem hiding this comment.
The tests I have added will then exactly catch that change and we can adapt it. Don't see the point in trying to anticipate the type of exception that this might throw in the future. Or did you mean to catch a more generic exception like OSError which is the base class of FileNotFoundError?
There was a problem hiding this comment.
right, I meant more generic.
Anyhow, if you think we should just use the one we have currently it's also ok. the tests will fail if this changes, so it will be fixed when it changes.
| @@ -71,7 +71,7 @@ def start(foreground, number): | |||
| subprocess.check_output(command, env=currenv, stderr=subprocess.STDOUT) # pylint: disable=unexpected-keyword-arg | |||
| except subprocess.CalledProcessError as exception: | |||
| click.secho('FAILED', fg='red', bold=True) | |||
| echo.echo_critical(exception.output) | |||
| echo.echo_critical(str(exception)) | |||
There was a problem hiding this comment.
Or use repr, see https://stackoverflow.com/a/4308203/1069467
There was a problem hiding this comment.
I used str since that is what is intended for normal users, whereas repr is for debugging purposes. I think in this case, showing the exception type for the user is not necessary. But if you think this would actually be better, happy to change it
|
good to go from me as well |
Fixes #3967
The commands
verdi node repo lsandverdi calcjob inputls/outputlswere passing the caught exception object straight into the echo call
without casting it to a string first, which would incur another
exception by itself. On top of that, they were catching
ValueErrorexceptions, which, as far as I can tell, would never be thrown. Instead
if the path does not exist a
FileNotFoundErrorwill be thrown. Thecommands now catch this instead and print a useful message. This failure
path of the commands was not tested, so the bug went unnoticed.
The problem of raw exception objects being passed to
echo_criticalwere present in some other commands as well, which have now been fixed.